how to scale `planned_analysis()` to multiple hypothesis/tests/models

aaronpeikert / repro-tutorial

https://doi.org/10.3390/psych3040053

Creative Commons Attribution 4.0 International

9 stars 4 forks source link

how to scale `planned_analysis()` to multiple hypothesis/tests/models #220

Closed aaronpeikert closed 3 years ago

cjvanlissa commented 3 years ago

Add an argument planned_analysis(which = "name") or planned_analysis(research_question = "name") or something similar?

cjvanlissa commented 3 years ago

If no argument is provided, it could just number them based on the metadata

brandmaier commented 3 years ago

I had the same question when I was reading this the first time. I understood this as a mere suggestion not necessarily a proposed standard? So it could be done in various other ways - but probably it's a good idea to give a suggestion of best practice in your view, @aaronpeikert .

aaronpeikert commented 3 years ago

Yes, it was merely a suggestion on how one could structure it. planned_analysis() is a user-written function but I expect most users prefer to write a script. Because many researchers number their hypothesis it would be most natural to name the functions/scripts according to this numbering scheme:

planned_analysis_1b()

planned_analysis_2_1.R

brandmaier commented 3 years ago

I suggest to clarify that this is not a genuine repro-function (like automate etc) but really just a suggestion how to structure your analysis in code.